microsoft/AI-For-Beginners

Public

mirrored fromhttps://github.com/microsoft/AI-For-BeginnersAvailable

CodeCommitsIssuesPull requestsActionsInsightsSecurity
16f27ee96997f339ca2245f299200f3df462dc45

Branches

Tags

  • No tags available.
0Branches0Tags
Go to file
Add file
Code

Clone

HTTPS

Download ZIP

lessons/4-ComputerVision/11-ObjectDetection/ObjectDetection.ipynb

602lines · modecode

1{
2 "cells": [
3 {
4 "cell_type": "markdown",
5 "metadata": {
6 "colab_type": "text",
7 "collapsed": true,
8 "id": "4Bta_vJzY-p2"
9 },
10 "source": [
11 "## Object Detection\n",
12 "\n",
13 "This is a notebooks from [AI for Beginners Curriculum](http://aka.ms/ai-beginners)"
14 ]
15 },
16 {
17 "cell_type": "markdown",
18 "metadata": {
19 "colab_type": "text",
20 "id": "H46ttrmEY-p4"
21 },
22 "source": [
23 "![Image Algorithms](https://cdn-images-1.medium.com/max/840/1*Hz6t-tokG1niaUfmcysusw.jpeg)"
24 ]
25 },
26 {
27 "cell_type": "markdown",
28 "metadata": {
29 "colab_type": "text",
30 "id": "1RDHP56XY-p6"
31 },
32 "source": [
33 "## Naive Approach to Object Detection\n",
34 "\n",
35 "* Break image into tiles\n",
36 "* Run CNN image classified through each time\n",
37 "* Select tiles with activation above the threshold"
38 ]
39 },
40 {
41 "cell_type": "code",
42 "execution_count": 11,
43 "metadata": {
44 "colab": {
45 "base_uri": "https://localhost:8080/",
46 "height": 35
47 },
48 "colab_type": "code",
49 "id": "jaR_CChGY-qN",
50 "outputId": "52b62ed4-6fd9-4e7c-827b-4e6457687e27",
51 "trusted": true
52 },
53 "outputs": [],
54 "source": [
55 "import cv2\n",
56 "from tensorflow import keras\n",
57 "import numpy as np\n",
58 "import matplotlib.pyplot as plt\n",
59 "import os"
60 ]
61 },
62 {
63 "cell_type": "markdown",
64 "metadata": {},
65 "source": [
66 "Let's read sample image to play with and pad it to square dimension:"
67 ]
68 },
69 {
70 "cell_type": "code",
71 "execution_count": 19,
72 "metadata": {
73 "colab": {
74 "base_uri": "https://localhost:8080/",
75 "height": 369
76 },
77 "colab_type": "code",
78 "id": "S0GTVupbY-qa",
79 "outputId": "99f192d3-570c-4e5a-d617-6992c5dc9098",
80 "trusted": true
81 },
82 "outputs": [
83 {
84 "data": {
85 "text/plain": [
86 "<matplotlib.image.AxesImage at 0x1cb8a16e8e0>"
87 ]
88 },
89 "execution_count": 19,
90 "metadata": {},
91 "output_type": "execute_result"
92 },
93 {
94 "data": {
95 "image/png": "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",
96 "text/plain": [
97 "<Figure size 432x288 with 1 Axes>"
98 ]
99 },
100 "metadata": {
101 "needs_background": "light"
102 },
103 "output_type": "display_data"
104 }
105 ],
106 "source": [
107 "img = cv2.imread('images/1200px-Girl_and_cat.jpg')\n",
108 "img = cv2.cvtColor(img,cv2.COLOR_BGR2RGB)\n",
109 "img = np.pad(img,((158,158),(0,0),(0,0)),mode='edge')\n",
110 "plt.imshow(img)"
111 ]
112 },
113 {
114 "cell_type": "markdown",
115 "metadata": {},
116 "source": [
117 "We will use pre-trained VGG-16 CNN:"
118 ]
119 },
120 {
121 "cell_type": "code",
122 "execution_count": 14,
123 "metadata": {
124 "colab": {
125 "base_uri": "https://localhost:8080/",
126 "height": 54
127 },
128 "colab_type": "code",
129 "id": "0RgUkBkchCxG",
130 "outputId": "1d5dc51e-3a19-43e8-8903-b85139f48fe6",
131 "trusted": false
132 },
133 "outputs": [],
134 "source": [
135 "vgg = keras.applications.vgg16.VGG16(weights='imagenet')"
136 ]
137 },
138 {
139 "cell_type": "markdown",
140 "metadata": {},
141 "source": [
142 "Let's define a function that will predict the probability of a cat on the image. Since ImageNet contains a number of classes for cats, indexed from 281 to 294, we will just add probabilities for those classes to get overall 'cat' probability:"
143 ]
144 },
145 {
146 "cell_type": "code",
147 "execution_count": 15,
148 "metadata": {
149 "colab": {
150 "base_uri": "https://localhost:8080/",
151 "height": 35
152 },
153 "colab_type": "code",
154 "id": "GzFYd9InhML4",
155 "outputId": "329354ce-5759-4236-e16b-d0c716893036",
156 "trusted": false
157 },
158 "outputs": [
159 {
160 "data": {
161 "text/plain": [
162 "0.61825"
163 ]
164 },
165 "execution_count": 15,
166 "metadata": {},
167 "output_type": "execute_result"
168 }
169 ],
170 "source": [
171 "def predict(img):\n",
172 " im = cv2.resize(img,(224,224))\n",
173 " im = keras.applications.vgg16.preprocess_input(im)\n",
174 " pr = vgg.predict(np.expand_dims(im,axis=0))[0]\n",
175 " return np.sum(pr[281:294]) # we know that VGG classes for cats are from 281 to 294\n",
176 "\n",
177 "predict(img)"
178 ]
179 },
180 {
181 "cell_type": "markdown",
182 "metadata": {},
183 "source": [
184 "Next function will build a heatmap of probabilities, dividing the image into $n\\times n$ squares:"
185 ]
186 },
187 {
188 "cell_type": "code",
189 "execution_count": 20,
190 "metadata": {
191 "colab": {
192 "base_uri": "https://localhost:8080/",
193 "height": 365
194 },
195 "colab_type": "code",
196 "id": "z5qqyj5uh6GN",
197 "outputId": "cf01a729-87e7-4b6e-8a43-89a61247df20",
198 "trusted": false
199 },
200 "outputs": [
201 {
202 "data": {
203 "text/plain": [
204 "<matplotlib.image.AxesImage at 0x1cb8983a940>"
205 ]
206 },
207 "execution_count": 20,
208 "metadata": {},
209 "output_type": "execute_result"
210 },
211 {
212 "data": {
213 "image/png": "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",
214 "text/plain": [
215 "<Figure size 1080x360 with 2 Axes>"
216 ]
217 },
218 "metadata": {
219 "needs_background": "light"
220 },
221 "output_type": "display_data"
222 }
223 ],
224 "source": [
225 "def predict_map(img,n):\n",
226 " dx = img.shape[0] // n\n",
227 " res = np.zeros((n,n),dtype=np.float32)\n",
228 " for i in range(n):\n",
229 " for j in range(n):\n",
230 " im = img[dx*i:dx*(i+1),dx*j:dx*(j+1)]\n",
231 " r = predict(im)\n",
232 " res[i,j] = r\n",
233 " return res\n",
234 "\n",
235 "fig,ax = plt.subplots(1,2,figsize=(15,5))\n",
236 "ax[1].imshow(img)\n",
237 "ax[0].imshow(predict_map(img,10))"
238 ]
239 },
240 {
241 "cell_type": "markdown",
242 "metadata": {},
243 "source": [
244 "## Detecting Simple Objects\n",
245 "\n",
246 "To give more precise location of a bounding box, we need to run **regression model** to predict bounding box coordinates. Let's start with simple example of having black rectangles in 32x32 images, which we want to detect. The idea and some code are borrowed from [this blog post](https://towardsdatascience.com/object-detection-with-neural-networks-a4e2c46b4491).\n",
247 "\n",
248 "The following function will generate a bunch of sample images:"
249 ]
250 },
251 {
252 "cell_type": "code",
253 "execution_count": 113,
254 "metadata": {
255 "colab": {},
256 "colab_type": "code",
257 "id": "KJSwfU3BGobM",
258 "trusted": false
259 },
260 "outputs": [
261 {
262 "name": "stdout",
263 "output_type": "stream",
264 "text": [
265 "Images shape = (100000, 8, 8)\n",
266 "BBoxes shape = (100000, 4)\n"
267 ]
268 }
269 ],
270 "source": [
271 "def generate_images(num_imgs, img_size=8, min_object_size = 1, max_object_size = 4):\n",
272 " bboxes = np.zeros((num_imgs, 4))\n",
273 " imgs = np.zeros((num_imgs, img_size, img_size)) # set background to 0\n",
274 "\n",
275 " for i_img in range(num_imgs):\n",
276 " w, h = np.random.randint(min_object_size, max_object_size, size=2)\n",
277 " x = np.random.randint(0, img_size - w)\n",
278 " y = np.random.randint(0, img_size - h)\n",
279 " imgs[i_img, x:x+w, y:y+h] = 1. # set rectangle to 1\n",
280 " bboxes[i_img] = [x, y, w, h]\n",
281 " return imgs, bboxes\n",
282 "\n",
283 "imgs, bboxes = generate_images(100000)\n",
284 "print(f\"Images shape = {imgs.shape}\")\n",
285 "print(f\"BBoxes shape = {bboxes.shape}\")"
286 ]
287 },
288 {
289 "cell_type": "markdown",
290 "metadata": {},
291 "source": [
292 "To make outputs of the network in the range [0;1], we will divide `bboxes` by the image size:"
293 ]
294 },
295 {
296 "cell_type": "code",
297 "execution_count": 114,
298 "metadata": {},
299 "outputs": [
300 {
301 "data": {
302 "text/plain": [
303 "array([0. , 0.25 , 0.125, 0.25 ])"
304 ]
305 },
306 "execution_count": 114,
307 "metadata": {},
308 "output_type": "execute_result"
309 }
310 ],
311 "source": [
312 "bb = bboxes/8.0\n",
313 "bb[0]"
314 ]
315 },
316 {
317 "cell_type": "markdown",
318 "metadata": {},
319 "source": [
320 "In our simple example, we will use dense neural network. In real life, when objects have more complex shape, it definitely makes sense to use CNNs for the task like this. We will use stochastic gradient descent optimizer and mean squared error (MSE) as the metrics, because our task is **regression**."
321 ]
322 },
323 {
324 "cell_type": "code",
325 "execution_count": 144,
326 "metadata": {},
327 "outputs": [
328 {
329 "name": "stdout",
330 "output_type": "stream",
331 "text": [
332 "Model: \"sequential_25\"\n",
333 "_________________________________________________________________\n",
334 " Layer (type) Output Shape Param # \n",
335 "=================================================================\n",
336 " flatten_24 (Flatten) (None, 64) 0 \n",
337 " \n",
338 " dense_50 (Dense) (None, 200) 13000 \n",
339 " \n",
340 " dropout_3 (Dropout) (None, 200) 0 \n",
341 " \n",
342 " dense_51 (Dense) (None, 4) 804 \n",
343 " \n",
344 "=================================================================\n",
345 "Total params: 13,804\n",
346 "Trainable params: 13,804\n",
347 "Non-trainable params: 0\n",
348 "_________________________________________________________________\n"
349 ]
350 }
351 ],
352 "source": [
353 "model = keras.Sequential([\n",
354 " keras.layers.Flatten(input_shape=(8,8)),\n",
355 " keras.layers.Dense(200, activation='relu'),\n",
356 " keras.layers.Dropout(0.2),\n",
357 " keras.layers.Dense(4)\n",
358 "])\n",
359 "model.compile('sgd','mse')\n",
360 "model.summary()"
361 ]
362 },
363 {
364 "cell_type": "markdown",
365 "metadata": {},
366 "source": [
367 "Let's train our network. We will also normalize the input data (by subtracting mean and dividing by standard deviation) for slightly better performance."
368 ]
369 },
370 {
371 "cell_type": "code",
372 "execution_count": 145,
373 "metadata": {},
374 "outputs": [
375 {
376 "name": "stdout",
377 "output_type": "stream",
378 "text": [
379 "Epoch 1/30\n",
380 "3125/3125 [==============================] - 6s 2ms/step - loss: 0.0562\n",
381 "Epoch 2/30\n",
382 "3125/3125 [==============================] - 5s 2ms/step - loss: 0.0131\n",
383 "Epoch 3/30\n",
384 "3125/3125 [==============================] - 5s 1ms/step - loss: 0.0076\n",
385 "Epoch 4/30\n",
386 "3125/3125 [==============================] - 5s 2ms/step - loss: 0.0058\n",
387 "Epoch 5/30\n",
388 "3125/3125 [==============================] - 5s 2ms/step - loss: 0.0050\n",
389 "Epoch 6/30\n",
390 "3125/3125 [==============================] - 5s 2ms/step - loss: 0.0044\n",
391 "Epoch 7/30\n",
392 "3125/3125 [==============================] - 5s 2ms/step - loss: 0.0041\n",
393 "Epoch 8/30\n",
394 "3125/3125 [==============================] - 5s 2ms/step - loss: 0.0038\n",
395 "Epoch 9/30\n",
396 "3125/3125 [==============================] - 5s 2ms/step - loss: 0.0036\n",
397 "Epoch 10/30\n",
398 "3125/3125 [==============================] - 6s 2ms/step - loss: 0.0034\n",
399 "Epoch 11/30\n",
400 "3125/3125 [==============================] - 8s 3ms/step - loss: 0.0033\n",
401 "Epoch 12/30\n",
402 "3125/3125 [==============================] - 8s 3ms/step - loss: 0.0031\n",
403 "Epoch 13/30\n",
404 "3125/3125 [==============================] - 6s 2ms/step - loss: 0.0030\n",
405 "Epoch 14/30\n",
406 "3125/3125 [==============================] - 6s 2ms/step - loss: 0.0029\n",
407 "Epoch 15/30\n",
408 "3125/3125 [==============================] - 5s 2ms/step - loss: 0.0028\n",
409 "Epoch 16/30\n",
410 "3125/3125 [==============================] - 5s 2ms/step - loss: 0.0028\n",
411 "Epoch 17/30\n",
412 "3125/3125 [==============================] - 5s 2ms/step - loss: 0.0027\n",
413 "Epoch 18/30\n",
414 "3125/3125 [==============================] - 5s 2ms/step - loss: 0.0026\n",
415 "Epoch 19/30\n",
416 "3125/3125 [==============================] - 5s 2ms/step - loss: 0.0026\n",
417 "Epoch 20/30\n",
418 "3125/3125 [==============================] - 5s 2ms/step - loss: 0.0025\n",
419 "Epoch 21/30\n",
420 "3125/3125 [==============================] - 6s 2ms/step - loss: 0.0024\n",
421 "Epoch 22/30\n",
422 "3125/3125 [==============================] - 6s 2ms/step - loss: 0.0024\n",
423 "Epoch 23/30\n",
424 "3125/3125 [==============================] - 6s 2ms/step - loss: 0.0024\n",
425 "Epoch 24/30\n",
426 "3125/3125 [==============================] - 6s 2ms/step - loss: 0.0023\n",
427 "Epoch 25/30\n",
428 "3125/3125 [==============================] - 5s 2ms/step - loss: 0.0023\n",
429 "Epoch 26/30\n",
430 "3125/3125 [==============================] - 5s 2ms/step - loss: 0.0022\n",
431 "Epoch 27/30\n",
432 "3125/3125 [==============================] - 5s 2ms/step - loss: 0.0022\n",
433 "Epoch 28/30\n",
434 "3125/3125 [==============================] - 5s 2ms/step - loss: 0.0022\n",
435 "Epoch 29/30\n",
436 "3125/3125 [==============================] - 5s 2ms/step - loss: 0.0021\n",
437 "Epoch 30/30\n",
438 "3125/3125 [==============================] - 5s 2ms/step - loss: 0.0021\n"
439 ]
440 },
441 {
442 "data": {
443 "text/plain": [
444 "<keras.callbacks.History at 0x1cbb7c499a0>"
445 ]
446 },
447 "execution_count": 145,
448 "metadata": {},
449 "output_type": "execute_result"
450 }
451 ],
452 "source": [
453 "imgs_norm = (imgs-np.mean(imgs))/np.std(imgs)\n",
454 "model.fit(imgs_norm,bb,epochs=30)"
455 ]
456 },
457 {
458 "cell_type": "markdown",
459 "metadata": {},
460 "source": [
461 "We seem to have relatively good loss, let's see how it translates into more tangible metrics, such as mAP. First, let's define IOU metric between two bounding boxes:"
462 ]
463 },
464 {
465 "cell_type": "code",
466 "execution_count": 146,
467 "metadata": {},
468 "outputs": [],
469 "source": [
470 "def IOU(bbox1, bbox2):\n",
471 " '''Calculate overlap between two bounding boxes [x, y, w, h] as the area of intersection over the area of unity'''\n",
472 " x1, y1, w1, h1 = bbox1[0], bbox1[1], bbox1[2], bbox1[3]\n",
473 " x2, y2, w2, h2 = bbox2[0], bbox2[1], bbox2[2], bbox2[3]\n",
474 " w_I = min(x1 + w1, x2 + w2) - max(x1, x2)\n",
475 " h_I = min(y1 + h1, y2 + h2) - max(y1, y2)\n",
476 " if w_I <= 0 or h_I <= 0: # no overlap\n",
477 " return 0.\n",
478 " I = w_I * h_I\n",
479 " U = w1 * h1 + w2 * h2 - I\n",
480 " return I / U"
481 ]
482 },
483 {
484 "cell_type": "markdown",
485 "metadata": {},
486 "source": [
487 "We will now generate 500 test images, and plot first 5 of them to visualize how accurate we are. We will print out IOU metric as well."
488 ]
489 },
490 {
491 "cell_type": "code",
492 "execution_count": 149,
493 "metadata": {},
494 "outputs": [
495 {
496 "name": "stdout",
497 "output_type": "stream",
498 "text": [
499 "pred=[3.7325673 3.6551285 2.0126944 1.030895 ],act=[4. 4. 2. 1.], IOU=0.41607480412565545\n",
500 "pred=[2.3762555 3.755858 1.1647941 1.0540264],act=[2. 4. 1. 1.], IOU=0.2932611042051458\n",
501 "pred=[-0.04900682 -0.10628867 2.7881489 1.027148 ],act=[0. 0. 3. 1.], IOU=0.7548650953478908\n",
502 "pred=[0.96806276 4.267275 1.3179774 1.1021365 ],act=[0. 5. 1. 1.], IOU=0.004833667961256898\n",
503 "pred=[2.157959 0.94667876 2.7568097 0.96259 ],act=[2. 1. 3. 1.], IOU=0.7965311752734529\n"
504 ]
505 },
506 {
507 "data": {
508 "image/png": "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",
509 "text/plain": [
510 "<Figure size 1080x360 with 5 Axes>"
511 ]
512 },
513 "metadata": {
514 "needs_background": "light"
515 },
516 "output_type": "display_data"
517 }
518 ],
519 "source": [
520 "import matplotlib\n",
521 "\n",
522 "test_imgs, test_bboxes = generate_images(500)\n",
523 "bb_res = model.predict((test_imgs-np.mean(imgs))/np.std(imgs))*8\n",
524 "\n",
525 "plt.figure(figsize=(15,5))\n",
526 "for i in range(5):\n",
527 " print(f\"pred={bb_res[i]},act={test_bboxes[i]}, IOU={IOU(bb_res[i],test_bboxes[i])}\")\n",
528 " plt.subplot(1,5,i+1)\n",
529 " plt.imshow(test_imgs[i])\n",
530 " plt.gca().add_patch(matplotlib.patches.Rectangle((bb_res[i,1],bb_res[i,0]),bb_res[i,3],bb_res[i,2],ec='r'))\n",
531 " #plt.annotate('IOU: {:.2f}'.format(IOU(bb_res[i],test_bboxes[i])),(bb_res[i,1],bb_res[i,0]+bb_res[i,3]),color='y')\n"
532 ]
533 },
534 {
535 "cell_type": "markdown",
536 "metadata": {},
537 "source": [
538 "Now to calculate mean precision over all cases, we just need to go over all our test samples, compute IoU, and calculate mean:"
539 ]
540 },
541 {
542 "cell_type": "code",
543 "execution_count": 151,
544 "metadata": {},
545 "outputs": [
546 {
547 "data": {
548 "text/plain": [
549 "0.7150587956049862"
550 ]
551 },
552 "execution_count": 151,
553 "metadata": {},
554 "output_type": "execute_result"
555 }
556 ],
557 "source": [
558 "np.array([IOU(a,b) for a,b in zip(test_bboxes,bb_res)]).mean()"
559 ]
560 },
561 {
562 "cell_type": "markdown",
563 "metadata": {},
564 "source": [
565 "## Real-Life Object Detection\n",
566 "\n",
567 "Real object detection algorithms are more complicated. We will recommend you to follow [Keras tutorial on Object Detection with RetinaNet](https://keras.io/examples/vision/retinanet/) if you want to get into the details of RetinaNet implementation, or use [Keras RetinaNet Library](https://github.com/fizyr/keras-retinanet), if you just want to train object detection model."
568 ]
569 }
570 ],
571 "metadata": {
572 "accelerator": "GPU",
573 "colab": {
574 "collapsed_sections": [],
575 "name": "ObjectDetection.ipynb",
576 "provenance": [],
577 "version": "0.3.2"
578 },
579 "interpreter": {
580 "hash": "16af2a8bbb083ea23e5e41c7f5787656b2ce26968575d8763f2c4b17f9cd711f"
581 },
582 "kernelspec": {
583 "display_name": "Python 3.8.12 ('py38')",
584 "language": "python",
585 "name": "python3"
586 },
587 "language_info": {
588 "codemirror_mode": {
589 "name": "ipython",
590 "version": 3
591 },
592 "file_extension": ".py",
593 "mimetype": "text/x-python",
594 "name": "python",
595 "nbconvert_exporter": "python",
596 "pygments_lexer": "ipython3",
597 "version": "3.8.12"
598 }
599 },
600 "nbformat": 4,
601 "nbformat_minor": 1
602}
603